A road network topology graph generation method and device, electronic equipment and medium

Abstract

The embodiment of the application provides a kind of road network topology graph generation method, device, electronic equipment and medium, with the following beneficial effects: obtaining trajectory data;According to the trajectory data, generate trajectory grid chart;According to trajectory data, the time sequence trajectory representation of each trajectory is obtained;According to trajectory grid chart, the trajectory grid representation of each clustering category is obtained;The similarity of the time sequence trajectory representation of each trajectory and trajectory grid representation is determined by the clustering category of each trajectory;The trajectory grid representation corresponding to the clustering category determined after fusion and the time sequence trajectory representation are obtained, and the fusion representation is obtained;According to the fusion representation, through model training, finally, the corresponding road network topology graph is generated.The image data and time sequence trajectory data represented by the trajectory of the application are fused, the representation ability of road network topology graph is enhanced, the precision is effectively improved, and the generation efficiency is improved by end-to-end generation method without additional algorithm.

Description

Technical Field

[0001] The present invention relates to the field of computers, and in particular to a method, device, electronic equipment and medium for generating a road network topology map. Background Art

[0002] The topological structure of the road network plays an important role in electronic navigation and traffic flow monitoring. In particular, the directionality and connectivity of the road network topological structure play a decisive role in the automatic path planning of the intelligent transportation system.

[0003] However, the existing road network topology map needs to be generated through a large number of user driving trajectories, and with the acceleration of urban development, road network information is usually constantly updated. Therefore, the original road network topology map generation method has the problems of being time-consuming and inaccurate.

[0004] Therefore, how to improve the efficiency and accuracy of road network topology map generation is a technical problem that technical personnel in this field urgently need to solve. Summary of the Invention

[0005] In order to overcome the shortcomings of low efficiency and poor accuracy of existing road network topology map generation methods, the present invention proposes a road network topology map generation method, device, electronic equipment and medium.

[0006] To achieve the above objectives, according to a first aspect of the present invention, an embodiment of the present invention provides a method for generating a road network topology map, the method comprising the following steps:

[0007] Get trajectory data;

[0008] generating a trajectory grid map according to the trajectory data;

[0009] Inputting the trajectory data into the Transformer model to obtain a temporal trajectory representation of each trajectory;

[0010] Inputting the trajectory grid graph into the VIT model, and obtaining the trajectory grid representation of each cluster category represented by the trajectory grid graph through MLP;

[0011] Calculating the similarity between the temporal trajectory representation and the trajectory grid representation of each trajectory, and determining the clustering category of each trajectory;

[0012] The trajectory grid representation and time series trajectory representation corresponding to the determined cluster category are fused to obtain the fused representation of each trajectory;

[0013] New trajectory data is generated through the Transformer model based on the fusion representation, and the corresponding road network topology map is generated by integrating the real grid map corresponding to the real road network map.

[0014] Optionally, generating a trajectory grid map according to the trajectory data includes:

[0015] Splitting the trajectory data into trajectory points of preset length;

[0016] Calculate the similarity of each trajectory point based on the Frechet distance, delete abnormal trajectory points and update the trajectory data;

[0017] Generate a trajectory grid map based on the updated trajectory data.

[0018] Optionally, the cluster categories include a straight trajectory category, a left-turn trajectory category, a right-turn trajectory category, or a combination of one or more of the following:

[0019] Optionally, fusing the trajectory grid representation corresponding to the determined cluster category with the time series trajectory representation to obtain a fused representation of each trajectory includes:

[0020] Add the trajectory grid representation and time series trajectory representation of the same cluster category to obtain the fused representation of each trajectory;

[0021] or,

[0022] The attention module is used to fuse the trajectory grid representation and time series trajectory representation of the same cluster category to obtain a fused representation of each trajectory.

[0023] According to a second aspect of the present invention, an embodiment of the present invention further provides a device for generating a road network topology map, comprising:

[0024] Acquisition module, used to obtain trajectory data;

[0025] A trajectory grid map generating module, configured to generate a trajectory grid map according to the trajectory data;

[0026] A temporal trajectory representation module, configured to input the trajectory data into a Transformer model to obtain a temporal trajectory representation of each trajectory;

[0027] A trajectory grid representation module is used to input the trajectory grid map into the VIT model and obtain the trajectory grid representation of each cluster category represented by the trajectory grid map through MLP;

[0028] A clustering module, configured to calculate the similarity between the temporal trajectory representation and the trajectory grid representation of each trajectory, and determine the clustering category of each trajectory;

[0029] A fusion representation module, configured to fuse the trajectory grid representation corresponding to the determined cluster category with the temporal trajectory representation to obtain a fusion representation of each trajectory;

[0030] The generation module is used to generate new trajectory data through the Transformer model according to the fusion representation, and integrate and generate the corresponding road network topology map after learning the real grid map corresponding to the real road network map.

[0031] Optionally, the trajectory grid map generating module is further configured to:

[0032] Splitting the trajectory data into trajectory points of preset length;

[0033] Calculate the similarity of each trajectory point based on the Frechet distance, delete the trajectory point and update the trajectory data;

[0034] Generate a trajectory grid map based on the updated trajectory data.

[0035] Optionally, the cluster categories include a straight trajectory category, a left-turn trajectory category, a right-turn trajectory category, or a combination of one or more of the following:

[0036] Optionally, the fusion characterization module is further used to:

[0037] Add the trajectory grid representation and time series trajectory representation of the same cluster category to obtain the fused representation of each trajectory;

[0038] or,

[0039] The attention module is used to fuse the trajectory grid representation and time series trajectory representation of the same cluster category to obtain a fused representation of each trajectory.

[0040] According to the third aspect of the present invention, an embodiment of the present invention further provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the road network topology map generation method in any of the above embodiments are implemented.

[0041] According to the fourth aspect of the present invention, an embodiment of the present invention also provides a storage medium, which stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, the at least one program, the code set or instruction set is loaded and executed by a processor to implement the steps of the road network topology map generation method in any of the above embodiments.

[0042] As described above, the embodiment of the present application provides a road network topology graph generation method, device, electronic equipment and medium, which has the following beneficial effects: trajectory data is obtained; a trajectory grid graph is generated according to the trajectory data; a time sequence trajectory representation of each trajectory is obtained by inputting the trajectory data into a Transformer model; a trajectory grid representation of each cluster category represented by the trajectory grid graph is obtained by inputting the trajectory grid graph into a VIT model and through an MLP; the cluster category of each trajectory is determined through the similarity of the time sequence trajectory representation and the trajectory grid representation of each trajectory; the trajectory grid representation corresponding to the determined cluster category and the time sequence trajectory representation are fused to obtain a fusion representation of each trajectory; new trajectory data is generated through a Transformer model according to the fusion representation, learned through a real grid map corresponding to a real road network graph, and integrated to generate a corresponding road network topology graph. The embodiment of the present application fuses the image data represented by the trajectory and the time sequence trajectory data through the model, thereby enhancing the representation ability of the road network topology graph, and effectively improving the precision. Moreover, the end-to-end generation method does not need to add an extraction algorithm, and further improves the generation efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0043] Figure 1 is a flowchart of a road network topology graph generation method provided by the embodiment of the present application;

[0044] Figure 2 is a flowchart of a trajectory grid graph generation method provided by the embodiment of the present application;

[0045] Figure 3 is a road network topology graph effect diagram provided by the embodiment of the present application;

[0046] Figure 4 is a structural diagram of a road network topology graph generation device provided by the embodiment of the present application;

[0047] Figure 5 is a hardware structure diagram of an electronic equipment for executing the road network topology graph generation method provided by the embodiment of the present application. DETAILED DESCRIPTION

[0048] In order for those skilled in the art to better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should belong to the protection scope of the present application.

[0049] Please refer to Figures 1 to 5It should be noted that the diagrams provided in this embodiment are merely schematic illustrations of the basic concept of the present invention. Therefore, the diagrams only show components related to the present invention and are not drawn according to the number, shape, and size of components in actual implementation. In actual implementation, the type, quantity, and proportion of each component may be changed arbitrarily, and the component layout may also be more complex.

[0050] See also Figure 1 , is a flow chart of a method for generating a road network topology diagram provided by an embodiment of the present invention, such as Figure 1 As shown, the embodiment of the present invention shows the process of the method for generating a road network topology map.

[0051] Step S101: Acquire trajectory data.

[0052] In specific implementation, the trajectory data can be collected by GPS (Global Positioning System) to collect continuous trajectory points, that is, τ = <(g1, t1), (g2, t2), ..., (g n ,t n )>, for each (g i ,t i ), i is a natural number, g i Represents each longitude and latitude coordinate value, t i The present invention generates a road network topology map of the area covered by the trajectory based on the above trajectory data, that is, G=(V, E), where V is a point in the road network and E is an edge of the road network.

[0053] Step S102: generating a trajectory grid diagram according to the trajectory data.

[0054] The trajectory data obtained in step S101 is used to further generate a trajectory grid map.

[0055] Since there may be noise in the trajectory data, in order to improve the accuracy of the trajectory grid map, see Figure 2 , is a flow chart of a method for generating a trajectory grid map provided by an embodiment of the present invention, such as Figure 2 As shown, the embodiment of the present invention shows the process of the trajectory grid map generation method.

[0056] Step S1021: splitting the trajectory data into trajectory points of preset length.

[0057] In an embodiment of the present invention, the trajectory points on a road can be split according to a preset length. Specifically, if the road includes 100 data points, the preset length can be 25, so that the data points on the road can be split into 4 parts.

[0058] Step S1022: Calculate the similarity of each trajectory point according to the frechet distance, delete the abnormal trajectory point and update the trajectory data.

[0059] After splitting the trajectory points, the present application uses the frechet distance to filter the trajectory points, specifically to calculate the similarity of each trajectory point. Further, those trajectory points with poor similarity represent a large degree of abnormal distance from the matched road, and are thus deleted as abnormal trajectory points. Finally, the trajectory data is updated, and the updated trajectory data completes data filtering to ensure the accuracy of the subsequent trajectory grid map.

[0060] Step S1023: Generate a trajectory grid map according to the updated trajectory data.

[0061] Screening accurate trajectory data T, which has a shape of [N, L, 2], where N is the number of trajectories, L is each trajectory τ, and generating a corresponding trajectory grid map according to the trajectory data.

[0062] Step S103: Input the trajectory data into the Transformer model to obtain the time sequence trajectory representation of each trajectory.

[0063] Input the trajectory data into the Transformer model to obtain the time sequence trajectory representation T of each trajectory t = Transformer(T). The Transformer model is a deep learning architecture that is trained through an encoder and a decoder to obtain trajectory representation, which is not described again in the embodiment of the present application.

[0064] Step S104: Input the trajectory grid map into the VIT model and obtain the trajectory grid representation of each cluster category represented by the trajectory grid map through the MLP.

[0065] VIT (English: Vision Transformer) is a deep learning method that uses the Transformer architecture to process image data. In the present application, the above trajectory grid map is input into the VIT model, and the trajectory grid representation of each cluster category represented by the trajectory grid map is obtained through the MLP (English: Multilayer Perceptron, Chinese: Multilayer Perceptron).

[0066] Imgk = Grid(Tk)

[0067] TG embedding = ViT(Imgk)

[0068] RT embedding = MLP(TG embedding)

[0069] The above shows the process of obtaining the trajectory grid representation, Img k represents the grid graph, T k is the trajectory data, ViT represents the VIT model, TG embedding represents the intermediate result after the VIT model is processed, and this TG embedding result is input into the MLP multi-layer perceptron to finally obtain the trajectory grid representation RT embedding.

[0070] In a specific implementation, the cluster category may include a straight trajectory category, a left-turn trajectory category, a right-turn trajectory category, or a combination of one or more of the following categories.

[0071] Step S105: Calculate the similarity between the temporal trajectory representation and the trajectory grid representation of each trajectory, and determine the cluster category of each trajectory.

[0072] After obtaining the temporal trajectory representation of each trajectory through the Transformer encoder, the cluster category to which the trajectory belongs is obtained by calculating the similarity with the road network trajectory representation.

[0073] Step S106: Fusing the trajectory grid representation and the time series trajectory representation corresponding to the determined cluster category to obtain a fused representation of each trajectory.

[0074] In an exemplary embodiment, in order to obtain a fused representation for each trajectory, the trajectory grid representation and the temporal trajectory representation of the same cluster category may be added together to obtain a fused representation for each trajectory.

[0075] In an exemplary embodiment, the attention module in the Transformer can also be used to fuse the trajectory grid representation and the time series trajectory representation of the same cluster category to obtain a fused representation of each trajectory.

[0076]

[0077] The process shown above realizes the optimization process of the fusion representation of the trajectory, where C j is the split trajectory area, T k is the trajectory data in the area, Transformer Encoder is the transformer model encoder, and the trajectory T k Mapped into its trajectory representation, TSC embedding i For each category representation of each cluster, m represents the number of fused road network representations in the area. As the network is updated and iterated (optimization function Loss), the trajectory representation will be mapped to the corresponding cluster representation TSCembedding i In this way, the representation of the road network trajectory is learned.

[0078] Step S107: Generate new trajectory data through the Transformer model based on the above fusion representation, and integrate and generate the corresponding road network topology map through learning the real grid map corresponding to the real road network map.

[0079]

[0080] The process shown above realizes the regeneration of multimodal fusion trajectory, where RT embedding is the trajectory grid representation within the region. i is the representation of each trajectory category of the fusion. i The road network trajectory generated by the model, It is represented as a fusion of the rasterized real road network trajectory representation and the representation of each trajectory category, which is matrix addition, to obtain the global trajectory cluster representation, and finally generate the road network trajectory points through the transformer decoder.

[0081] Troad_map={(f(xi,yi))|(xi,yi)∈Tk}

[0082]

[0083] The process shown above realizes the learning process of the generated road network trajectory and the real road network, where T k is the actual trajectory point, including longitude and latitude (x i ,y i ), T road_map is a road network trajectory point that passes through the coordinate point (u i ,u j ) are matched with the real trajectory points to select the matching road network trajectory points. Finally, the generated trajectory points are aligned with the road network trajectory points through the optimization function Loss (n is the number of trajectory points in the area) to determine the accuracy of the road network representation.

[0084] The obtained trajectory cluster representation is decoded to generate the corresponding trajectory data, and the generated trajectory data is obtained by learning with the real grid map generated by the real road network map. Finally, the corresponding road network map is generated by integrating the trajectory data.

[0085] See also Figure 3, is a schematic diagram of the road network topology effect provided by an embodiment of the present invention, in which respectively the real road network map corresponding to the trajectory, i.e., the true value, the picture corresponding to the trajectory data, i.e., the trajectory point, the generated raster road network map, and the road network topology map generated by the present invention are shown. The road network topology map has a high degree of consistency with the real road network map, reflecting the accuracy of the road network topology map generated by the present invention.

[0086] As can be seen from the description of the above embodiments, an embodiment of the present invention provides a method for generating a road network topology map, comprising obtaining trajectory data; generating a trajectory grid map based on the trajectory data; obtaining a time-series trajectory representation of each trajectory based on the trajectory data; obtaining a trajectory grid representation of each cluster category based on the trajectory grid map; determining the cluster category of each trajectory based on the similarity between the time-series trajectory representation and the trajectory grid representation of each trajectory; fusing the trajectory grid representation corresponding to the determined cluster category with the time-series trajectory representation to obtain a fused representation; and finally generating a corresponding road network topology map through model training based on the fused representation. The present invention fuses the image data represented by the trajectory and the time-series trajectory data, thereby enhancing the representation capability of the road network topology map and effectively improving the accuracy. Moreover, the end-to-end generation method does not require additional algorithms, thereby improving the generation efficiency.

[0087] Through the description of the above method embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software plus a necessary general hardware platform, or of course by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk.

[0088] An embodiment of the present invention provides a non-volatile computer storage medium, wherein the computer storage medium stores computer-executable instructions, and the computer-executable instructions can execute the road network topology map generation method in any of the above method embodiments.

[0089] Corresponding to the embodiment of the battery health status estimation method provided by the present invention, the present invention also provides a road network topology map generating device.

[0090] See also Figure 4 , is a schematic diagram of the structure of a road network topology map generating device provided by an embodiment of the present invention. As shown in the figure, the device includes:

[0091] An acquisition module 11 is used to acquire trajectory data;

[0092] A trajectory grid map generating module 12 is configured to generate a trajectory grid map based on the trajectory data;

[0093] A time series trajectory representation module 13 is used to input the trajectory data into a Transformer model to obtain a time series trajectory representation of each trajectory;

[0094] A trajectory grid representation module 14 is configured to input the trajectory data into a VIT model and obtain a trajectory grid representation of each cluster category represented by the trajectory grid graph through MLP;

[0095] Clustering module 15, used to calculate the similarity between the time series trajectory representation and the trajectory grid representation of each trajectory, and determine the cluster category of each trajectory;

[0096] A fusion representation module 16 is configured to fuse the trajectory grid representation corresponding to the determined cluster category with the time series trajectory representation to obtain a fusion representation of each trajectory;

[0097] The generation module 17 is used to generate new trajectory data through the Transformer model according to the fusion representation, and integrate and generate the corresponding road network topology map after learning the real grid map corresponding to the real road network map.

[0098] In an exemplary embodiment, the trajectory grid map generating module 12 is further configured to:

[0099] Splitting the trajectory data into trajectory points of preset length;

[0100] Calculate the similarity of each trajectory point based on the Frechet distance, delete the trajectory point and update the trajectory data;

[0101] Generate a trajectory grid map based on the updated trajectory data.

[0102] In an exemplary embodiment, the cluster categories include a combination of one or more of a straight trajectory category, a left-turn trajectory category, and a right-turn trajectory category.

[0103] In an exemplary embodiment, the fusion characterization module 16 is further configured to:

[0104] Add the trajectory grid representation and time series trajectory representation of the same cluster category to obtain the fused representation of each trajectory;

[0105] or,

[0106] The attention module is used to fuse the trajectory grid representation and time series trajectory representation of the same cluster category to obtain a fused representation of each trajectory.

[0107] Figure 5 FIG is a schematic diagram of the hardware structure of an electronic device for executing a method for generating a road network topology map according to an embodiment of the present invention. Figure 5 As shown, the device includes:

[0108] One or more processors 510 and memory 420, Figure 5 A processor 510 is taken as an example.

[0109] The device for executing the method for generating a road network topology map may further include: an input device 530 and an output device 540 .

[0110] The processor 510, the memory 520, the input device 530 and the output device 540 may be connected via a bus or other means. Figure 5 The bus connection is taken as an example.

[0111] The memory 520 is a non-volatile computer-readable storage medium that can be used to store non-volatile software programs, non-volatile computer executable programs and modules, such as the program instructions / modules corresponding to the road network topology map generation method in the embodiment of the present invention (for example, the attached Figure 4 The processor 510 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions, and modules stored in the memory 520, thereby implementing the road network topology map generation method of the above method embodiment.

[0112] The memory 520 may include a program storage area and a data storage area. The program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created by the processing device of the road network map generation method. Furthermore, the memory 520 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state memory device. In some embodiments, the memory 520 may optionally include a memory remotely located relative to the processor 510. These remote memories may be connected to the processing device of the road network topology map generation method via a network. Examples of the aforementioned networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0113] The input device 530 can receive input digital or character information and generate key signal input related to user settings and function control of the processing device for generating the road network topology map. The output device 540 can include a display device such as a display screen.

[0114] The one or more modules are stored in the memory 520 and, when executed by the one or more processors 510, perform the road network topology map generating method in any of the above method embodiments.

[0115] The above-mentioned product can execute the method provided by the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not fully described in this embodiment, please refer to the method provided by the embodiment of the present invention.

[0116] The electronic devices according to the embodiments of the present invention may be implemented in various forms, including but not limited to:

[0117] (1) Mobile communication devices: These devices are characterized by their mobile communication capabilities and are primarily designed to provide voice and data communications. These terminals include smartphones (e.g., iPhones), multimedia phones, feature phones, and low-end phones.

[0118] (2) Ultra-mobile personal computer devices: These devices fall under the category of personal computers, have computing and processing capabilities, and generally also have mobile Internet access. These terminals include PDAs, MIDs, and UMPCs, such as the iPad.

[0119] (3) Portable entertainment devices: These devices can display and play multimedia content. These devices include audio and video players (such as iPods), handheld game consoles, e-books, smart toys, and portable car navigation devices.

[0120] (4) Server: A device that provides computing services. The server consists of a processor, hard disk, memory, system bus, etc. The server is similar to a general computer architecture, but because it needs to provide highly reliable services, it has higher requirements in terms of processing power, stability, reliability, security, scalability, and manageability.

[0121] (5) Other electronic devices with data interaction functions.

[0122] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of this embodiment.

[0123] Each embodiment in this specification is described in a progressive manner. The same or similar parts between the embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the device or system embodiments, since they are basically similar to the method embodiments, the description is relatively simple. For the relevant parts, refer to the partial description of the method embodiments. The device and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. A person of ordinary skill in the art can understand and implement it without making any creative efforts.

[0124] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0125] The foregoing description is intended only to provide specific embodiments of the present invention, which will enable those skilled in the art to understand and implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown herein, but is to be embodied in the widest possible manner consistent with the principles and novel features disclosed herein.

Claims

1. A method for generating a road network topology map, characterized in that: include: Get trajectory data; generating a trajectory grid map according to the trajectory data; Inputting the trajectory data into the Transformer model to obtain a temporal trajectory representation of each trajectory; Inputting the trajectory grid graph into the VIT model, and obtaining the trajectory grid representation of each cluster category represented by the trajectory grid graph through MLP; Calculating the similarity between the temporal trajectory representation and the trajectory grid representation of each trajectory, and determining the clustering category of each trajectory; The trajectory grid representation and the time series trajectory representation corresponding to the determined cluster category are fused to obtain a fused representation of each trajectory, including: adding the trajectory grid representation and the time series trajectory representation of the same cluster category to obtain a fused representation of each trajectory; or using the attention module to fuse the trajectory grid representation and the time series trajectory representation of the same cluster category to obtain a fused representation of each trajectory; New trajectory data is generated through the Transformer model based on the fusion representation, and the corresponding road network topology map is generated by integrating the real grid map corresponding to the real road network map.

2. The method for generating a road network topology map according to claim 1, wherein: Generating a trajectory grid map according to the trajectory data includes: Splitting the trajectory data into trajectory points of preset length; Calculate the similarity of each trajectory point based on the Frechet distance, delete abnormal trajectory points and update the trajectory data; Generate a trajectory grid map based on the updated trajectory data.

3. The method for generating a road network topology map according to claim 1 or 2, wherein: The cluster categories include a straight trajectory category, a left-turn trajectory category, a right-turn trajectory category, or a combination of one or more of the following:

4. A road network topology map generating device, characterized in that: include: Acquisition module, used to obtain trajectory data; A trajectory grid map generating module, configured to generate a trajectory grid map according to the trajectory data; A temporal trajectory representation module, configured to input the trajectory data into a Transformer model to obtain a temporal trajectory representation of each trajectory; A trajectory grid representation module is used to input the trajectory grid map into the VIT model and obtain the trajectory grid representation of each cluster category represented by the trajectory grid map through MLP; A clustering module, configured to calculate the similarity between the temporal trajectory representation and the trajectory grid representation of each trajectory, and determine the clustering category of each trajectory; A fusion representation module, configured to fuse the trajectory grid representation corresponding to the determined cluster category with the temporal trajectory representation to obtain a fusion representation of each trajectory; The fusion representation module is further used to: add the trajectory grid representation and the time series trajectory representation of the same cluster category to obtain a fusion representation of each trajectory; or use the attention module to fuse the trajectory grid representation and the time series trajectory representation of the same cluster category to obtain a fusion representation of each trajectory; The generation module is used to generate new trajectory data through the Transformer model according to the fusion representation, and integrate and generate the corresponding road network topology map after learning the real grid map corresponding to the real road network map.

5. The road network topology map generating device according to claim 4, characterized in that: The trajectory grid map generation module is further used to: Splitting the trajectory data into trajectory points of preset length; Calculate the similarity of each trajectory point based on the Frechet distance, delete the trajectory point and update the trajectory data; Generate a trajectory grid map based on the updated trajectory data.

6. The road network topology map generating device according to claim 4 or 5, characterized in that: The cluster categories include a straight trajectory category, a left-turn trajectory category, a right-turn trajectory category, or a combination of one or more of the following:

7. An electronic device, characterized in that: The method comprises a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps of the method for generating a road network topology map according to any one of claims 1 to 3 are implemented.

8. A computer-readable storage medium, characterized in that The storage medium stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, the at least one program, the code set or instruction set is loaded and executed by the processor to implement the steps of the road network topology map generation method as described in any one of claims 1 to 3.

Images